Asymptotic Theory for Weighted Least Squares Estimators in Aalen's Additive Risk Model

摘要

Let h(t/Z sub i) be the conditional hazard function for the survival time of an individual sub i given the p-dimensional covariate process Z sub i(t). This document inference for Aalen's additive risk model h(t/Z sub i)=Z sub i(t) alpha (t), where alpha is a p-vector of unknown hazard functions. The theory of counting processes is used to obtain weak convergence results for weighted least squares estimators of the hazard functions and the cumulative hazard functions based on continuous data. Results for weighted least squares estimators based on grouped data are also described. Keywords: Regression models, Biostatistics.